Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at IFTO-INSTITUTO FEDERAL DE TOCANTINS on January 26, 2016. For personal use only. No other uses without permission. Copyright © 2001 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
Journal of Orthopaedic & Sports Physical Therapy 2OOl;3l (ll):620-631
Design and Implementation of a ~eu~omuscular Training Program Following Anterior Cruciate Ligament ~econstruction May Arna Risberg, P7; PhD Marianne Msrk, PTZ Hanne Krogstad )enssen, PTZ lnger Holm, P7; PhD3
Neuromuscular training programs are increasingly integrated into clinical practice for lower extremity rehabilitation. A few rehabilitation programs have been evaluated for patients with anterior cruciate ligament (ACL) deficiency and for injury prevention, but there is limited scientific evidence of the effect of neuromuscular training following ACL reconstruction. Therefore, a neuromuscular training program was developed for patients after ACL reconstruction. The objective of the neuromuscular training was to improve the ability to generate a fast and optimal muscle firing pattern, to increase dynamic joint stability, and to relearn movement patterns and skills necessary during activities of daily living and sports activities. The main areas considered when designing the postoperative rehabilitation program after ACL reconstruction were: ACL graft healing and ACL strain values during exercises, proprioception and neuromuscular control, and clinical studies on the effect of neuromuscular training programs. The rehabilitation program consists of balance exercises, dynamic joint stability exercises, jump traininglplyometric exercises, agility drills, and sportspecific exercise. The patients exercise 3 times a week for 6 months. The scientific and clinical evidence for the rehabilitation program are described and the main exercises in the program are outlined. / Orthop Sports Phys Ther 200 1;3 1 :62O-631.
Key Words: dynamic joint control, knee surgery, proprioception
euromuscular training has become integrated into clinical practice for both upper and lower extremity rehabilit a t i ~ n .According ~ ~ . ~ ~ to the definition of neuromuscular control, neuromuscular training could be defined as training enhancing unconscious motor responses by stimulating both afferent signals and central mechanisms responsible for dynamic joint control. The biomechanics of the knee are altered after anterior cruciate ligament (ACL) injury, but neuromuscular training may
enhance control of abnormal joint translation during functional activities by inducing compensatory alterations in muscle activity patterns. The objectives of neuromuscular training are to improve the nervous system's ability to generate a fast and optimal muscle firing pattern, to increase dynamic joint stability, to decrease joint forces, and to relearn movement patterns and skills. The exercises are designed to induce compensatory changes in muscle activation patterns and facilitate dynamic joint stability in patients with ACL injury. The goal is to achieve a state of "readiness" of muscles to respond to joint forces resulting in enhanced motor ~ontrol.".~" Several studies have evaluated outcome after ACL reconstruction,2.6.'>2~2H..90..92.4".41.47.34.74.73 but
very few have evaluated the effect of different rehabilitation programs following ACL Only a few studies have evaluated the effect of neuromuscular training, and most of those have foCenter for Clinical Research, Ullevaal University Hospital, Oslo, Norway. cused on either subjects with ACL Department for Physical and Rehabilitation Medicine, Ullevaal University Hospital, Oslo, Norway. deficient knee^^:'^.^^ or the effect Biomechanics laboratory, National Hospital Orthopaedic Center, University of Oslo, Norway. Send correspondence to May Arm Risberg, Center for Clinical Research, Ullevaal University Hos- of training on injury prevention. lH,.36.37.71 pital, 0407, Oslo, Norway. E-mail:
[email protected]
Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at IFTO-INSTITUTO FEDERAL DE TOCANTINS on January 26, 2016. For personal use only. No other uses without permission. Copyright © 2001 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
Ihara and N a k a ~ a m awere ~ ~ the first to evaluate neuromuscular training consisting of balance exercise and perturbation training. Four subjects with ACMeficient knees who had "giving way" symptoms went through a training program 4 times per week for 3 months. The patients were compared to a control group of 5 subjects. Significant improvement was found in peak torque time and rising torque value of the hamstrings in the training group compared to the control group. These researchers concluded that the training program apparently had the potential to shorten the time lag of muscular reaction. Beard et alRstudied 50 subjects with ACMeficient knees randomly assigned to a proprioceptive training program or a traditional strength training program. The proprioceptive program included balance, dynamic joint stability, and perturbation training. Both programs were performed using circuit training. Warm-up and stretching preceded and followed the exercise circuit. The neuromuscular training program consisted of 1 hour of intensive training 2 times per week for 12 weeks and a home exercise program. The exercises were performed in weightbearing positions. The traditional strength training program included the same number of training sessions, but the exercises were mostly performed in non-weight-bearing positions with the objective of increasing the strength of lower limb muscles. No attempt was made to increase speed of contraction. An indirect measurement of proprioception, the reflex hamstring contraction latency, was used in addition to Lysholm functional score and knee joint laxity. The study demonstrated a significant improvement in the neuromuscular training group for mean hamstring contraction latency and for Lysholm functional score compared to the traditional strength training group. studied 26 patients with acute Fitzgerald et ACL rupture. The patients were randomly assigned to a standard rehabilitation program with or without perturbation training exercises. Adding the perturbation training to the standard rehabilitation program reduced the risk of continued episodes of giving way of the knee during sports activities and maintained the knee function at the 6-month follow-up compared to the patients in the standard rehabilitation program. Hewett et al" developed a jump/plyometric training program to evaluate the effect of the training program on landing mechanics and lower extremity strength in female athletes. The program was designed to decrease landing forces by teaching neuromuscular control during landing and to increase vertical jump height. The same group did a prospective study where a jump training program was included to reduce the risk of knee injuries in athletes.% Their study showed that the incidence of knee injuJ Orthop Sports Phys Therevolume 31 .Number 11 .November 2001
ries decreased significantly in female athletes compared to male athletes.36 To our knowledge, no studies have been published on the effect of neuromuscular training after ACL re; construction. Current research on the effect of neuromuscular training, knowledge about graft healing after ACL reconstruction (bone-patellar tendon-bone graft), research on proprioception and neuromuscular control, and our clinical experience with patients who have ACL reconstruction were considered during the design of our rehabilitation p r ~ g r a m . ~ "
EVIDENCE GUIDING THE DEVELOPMENT OF THE NEUROMUSCULAR REHABILITATION PROGRAM FOLLOWING ANTERIOR CRUCIATE LIGAMENT RECONSTRUCTION The main topics for developing the neuromuscular rehabilitation program following ACL reconstruction involve: (1) graft healing response and ACL strain values during exercises, (2) function of ligament mechanoreceptors, and (3) neuromuscular control. However, evidence in this field is limited and inconclusive, and there are many conflicting and controversial areas. Following our discussion of the 3 main topics, the different types of training included in the rehabilitation program are described, and the main exercises for each type of training are outlined. The entire rehabilitation program is described in the A p pendix.
Graft Healing Response and Anterior Cruciate Ligament Strain Values During Exercises The goal of the rehabilitation program after ACL reconstruction is to restore joint motion and lower limb performance, and regain muscle strength to the preinjury level without producing excessive graft strain during healing of the graft. In contrast to development of a rehabilitation program for patients with ACL deficiency, the graft healing responses and ACL strain values during exercises have to be considered in the design of a rehabilitation program for patients after ACL reconstruction. Experimental studies of healing of the ACL graft have demonstrated that the graft requires a long time to revascularize and heal.'-J5Therefore, the ACL graft has been considered vulnerable during the first to reweeks after ACL reconstruction.1~4~ccording sults from animal studies, strenuous rehabilitation exercises should not be included immediately after surgery, and aggressive rehabilitation programs should not start immediately after ACL r e c o n s t r ~ c t i o n . ' ~ ~ ~ However, experience from use of graft materials with biomechanical properties similar to the normal ACL,'* adequate fixation ~trength;~' and clinical evidence of no increased laxity after aggressive rehabili62 1
Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at IFTO-INSTITUTO FEDERAL DE TOCANTINS on January 26, 2016. For personal use only. No other uses without permission. Copyright © 2001 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
position as well as for initiating reflex muscle contation program^^^.^^^^^ have led many to recommend traction about the knee.s1 aggressive rehabilitation programs. But evidence for The decrease of proprioceptive sense in the ACL the remodeling process in the human ACL graft is .~~.~~.~~ unknown. A previous study by our group demonstrat- deficient knee has been well e ~ a m i n e d . ~ Bared that linear stiffness properties and ultimate failure rack et a14 demonstrated a significant decrease in kinesthesia of 37% in patients after complete ACL load of a human ACL graft 8 months after reconrupture and concluded that patients who sustained a struction were about 90% of the values for the unincomplete ACL tear may experience a decline in projured knee.13 That case study indicated that strenuprioceptive function contributing to the progressive ous exercises could at least be introduced 8 months instability and disability often observed after this inafter ACL reconstruction. jury. Moreover, Corrigan et alZ0and MacDonald et The effect of different rehabilitation exercises on a14Rreported a significant increase of 39% and 21%. the healing response of an ACL graft is still unclear. respectively, in the threshold to detection of passive However, different strain values of the ACL have motion in ACMeficient knees. Beynnon et all4 also been reported for different rehabilitation exercisreported a significant decrease in kinesthetic awarees.lOJ1.m Beynnon et all1 studied the effect of weightness in ACLdeficient knees. bearing and non-weight-bearing exercises on strain Different results of the effect of ACL surgery on behavior of the normal ACL in human subjects while proprioception have been r e p ~ r t e d . ~ . MacDon~~.~"~~ performing squatting and active flexionextension of ald et a14%tudied proprioception in subjects with the leg. They showed that maximum ACL strain valACL reconstruction and found that proprioception ues obtained during squatting did not differ from did not improve after reconstruction when compared those obtained during active flexionextension. Furthermore, the strain values during squatting were un- to subjects with ACMeficient knees. Barrett5 studied joint position sense in subjects with ACLdeficient, affected by the application of elastic resistance durACL-reconstructed, and normal knees and reported ing the exercise. These data indicate that nonsignificantly poorer joint position sense in ACL-reweight-bearing and weight-bearing exercises produce constructed knees compared to normal knees, but about the same amount of strain, suggesting that significant improvement compared to ACMeficient non-weight-bearing exercises can be introduced as knees. Lephart et examined patients with ACL early as weight-bearing exercises after ACL reconreconstruction 11 to 26 months after surgery and obstruction. However, weight-bearing exercises have served a significant increase in threshold to detecbeen shown to produce a more favorable functional tion of passive motion at near extension in reconoutcome and are usually recommended." structed knees. These results were confirmed by a reBeynnon et all2 evaluated the effect of an aggressive versus a nonaggressive rehabilitation p r ~ g r a m . ~ cent study by Fischer-Rasmussen and Jensen,2%ho found decreased proprioception of the knee joint in These rehabilitation programs were designed based on their previous studies of ACL strain values for dif- people with both ACMeficient and ACL-reconstructed knees. Fremerey et alw recently published a proferent rehabilitation exercises. The program includspective longitudinal study evaluating proprioception ed the same exercises, but one group started the exercises at week 2 and the other group at week 6. Pre- (joint position sense) after ACL reconstruction and found impaired proprioception 3 months after surliminary results indicated that prescribing exercises gery compared with preoperative findings. Six early after ACL reconstruction (accelerated rehabilimonths postoperatively, the proprioception was retation program) may increase knee joint laxity, but stored to near full extension and flexion, whereas no significant differences were found regarding paproprioception in the midrange position was still imtient perception of knee function. Other nonranpaired. A strong correlation ( r = 0.76) was found bedomized clinical trials on aggressive rehabilitation programs have concluded that these programs are ef- tween proprioception and patient satisfaction. Results from studies evaluating proprioceptive defifective and d o not produce increased knee joint laxicits after ACL reconstruction vary. This can partly be ty.63 Based on these conflicting results, more eviexplained by the recent evidence of the timing of dence from clinical randomized trials is needed. reinnervation of the graft (postoperatively) and the wide variety of tests used to evaluate proprioception. Function of Ligament Mechanoreceptors It has been demonstrated that free patellar tendon grafts in dogs are partly reinnervated 6 months after Anatomical studies have demonstrated the exissurgery.Wthough no innervation was observed in tence of mechanoreceptors in the human ACL.'j".61 the ACL grafts immediately after surgery, histological Pitman et aP2 used arthroscopic procedures to proexamination of the graft tissue 6 months after survide direct evidence for the presence of active progery revealed that all 6 grafts contained neural eleprioceptive receptors within the intact ACL of the ments with equal numbers of mechanoreceptors and human knee. Animal studies have shown that these free nerve endings. Other animal studies have conparticular receptors are specific for detecting joint J Orthop Sports Phys Ther.Volume 31 .Number 1 1 .November 2001
Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at IFTO-INSTITUTO FEDERAL DE TOCANTINS on January 26, 2016. For personal use only. No other uses without permission. Copyright © 2001 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
firmed that mechanoreceptors regenerate gradually after ACL r e c o n ~ t r u c t i o nHuman .~ studies have confirmed the results of these animal studies. Ochi et alm found that sensory reinnervation in the reconstructed ACL was closely related to the function of the knee. Barrett3 identified the recovery of proprioception as the most important factor for a functionally good result after ACL reconstruction. Co et all9 reported a significant difference in proprioception between ACLreconstructed and normal knees but concluded that the results were not clinically significant because the reconstructed and the normal knees both performed better than the normal control knees. They suggested that training could restore much of the loss in proprioception because the subjects with reconstruction went through a rehabilitation program for both knees. The contribution of joint receptors to kinesthesia and position sense has been a matter of debate, and the topic has frequently been revie~ed.~' Evidence exists for a significant influence of muscle afferents and ligamentous joint afferents on position and movement sense. There are also reasons to believe that there might be considerable variation among different individuals in the use of joint afferent inform a t i ~ nand , ~ there might be inherent differences among subjects. Possible genetic differences among different individuals have been studied.39This research has included an assessment of the subjects' abilities and enduring characteristics or traits."Vhe "single subject" designs differ from the experimental approach in which differences among individuals are ignored in order to concentrate on the average performances of larger groups of subjects. More research on individual differences would contribute to the knowledge of the effect of rehabilitation and training programs.
Neuromuscular Control Simply restoring mechanical restraints is not enough for a functional recovery of the kneeM,".% because the coordinated neuromuscular controlling mechanism required during daily living and sportspecific activities would be neglected. Rehabilitation programs cannot alter a mechanical knee joint instability but may affect the neuromuscular control and the dynamic joint stability. A lag in the neuromuscular reaction time can result in dynamic joint instability with recurrent episodes of joint subluxation and deterioration. Therefore, both mechanical stability and neuromuscular control are probably important for long-term functional outcome, and both aspects must be considered in the design of a neuromuscular rehabilitation program after ACL reconstruction. Injury to the ACL has been shown to result in altered somatosensory information that may adversely J Orthop Sports Phys Ther.Volume 31 .Number 11 .November 2001
affect neuromuscular control.55Several studies have reported differences in neuromuscular performance in ACMeficient and reconstructed knee~.'-~"eard at a17 studied the reflex hamstring contraction latency in subjects with reconstruction preoperatively and then postoperatively at 3 and 6 months. There was a significant improvement in the neuromuscular control following reconstruction of the ACL, but only for those with poor proprioception prior to surgery. Huston and WojtysSRevaluated possible predisposing neuromuscular factors for ACL injuries in athletes of both sexes compared to nonathletes. Female athletes required significantly more time to generate maximum hamstring muscle torque than male athletes. A different muscle recruitment order was found in a limited subgroup of female athletes. Furthermore, significant differences in the recruitment order of the lower extremity muscles were observed between males and females. In a recent study, Wojtys and Huston7' found significantly impaired neuromuscular function 12 to 18 months after ACL reconstruction, although 80% of the subjects believed that they had regained their preoperative functional level. Quadriceps and hamstring muscle reaction time were identified as the best indicators of subjective knee function. Some patients with ACL injuries compensated well for their loss of the knee stabilizer (copers), but others did not ( n o n c o p e r ~ ) .Rudolph ~ . ~ ~ et a15%dentified movement strategies in copers and noncopers and found that copers had movement strategies similar to those of uninjured subjects. Copers stabilized their knee with a greater contribution from the ankle plantar flexors, indicating that the significance of gastrocnemius muscles as contributors to dynamic knee joint stability should be emphasized during the rehabilitation program. The noncopers showed a reduction in range of motion and external knee flexion moment, but no evidence of quadriceps avoidance gait was found. Neither copers nor noncopers had any reduction in quadriceps activation.
THE NEUROMUSCULAR REHABILITATION PROGRAM The rehabilitation program consists of balance exercises, dynamic joint stability exercises, plyometric exercises, agility drills, and sport-specific exercises. The program is divided into 6 phases of 3 to 5 weeks each (Appendix). Specific exercises are described for each week in the rehabilitation protocol. Most patients in our rehabilitation program are ready to progress at the speed given in the protocol, but not all patients will be able to progress at the same pace. Patients who sustain pain, swelling, or range of motion deficits undergo treatments until these impairments are resolved. Criteria used to determine readiness for progression include no increased pain or
Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at IFTO-INSTITUTO FEDERAL DE TOCANTINS on January 26, 2016. For personal use only. No other uses without permission. Copyright © 2001 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
FIGURE 1. Dynamic joint stability exercise using the "star."
swelling and the ability to maintain postural control of the position before movements are superimposed on the position. The patients first need to be aware of the position of the body in space before tolerating movements into space or reacting to a perturbation. When the patient is able to successfully perform the
exercise on a flat, even surface, the exercise is made more challenging by changing the surface to balance mats, a wobble board, or a trampoline. Furthermore, sensory feedback is challenged by excluding vision, challenging the vestibular system through changing the base of support, and using distractions, such as a
I
?
FIGURE 2. (A) Balance reach leg. 624
I
(B)Balance reach arm. J Orthop Sports Phys Ther.Volume 31 .Number I l *November 2001
Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at IFTO-INSTITUTO FEDERAL DE TOCANTINS on January 26, 2016. For personal use only. No other uses without permission. Copyright © 2001 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
"ti
FIGURE 3. Lunge exercise using the "star."
ball or sudden and unexpected change in movement directions. No knee braces are used following the knee surgery o r during the rehabilitation program.
1,. ;i--Tyr?..'..-. -..... . .,-. .- ......:.I
-
"
Berg9 described "balance" as the ability to maintain a position, the ability to voluntarily move, and the ability to react to a perturbation. All of these components are important for neuromuscular control. To distinguish between balance training, dynamic joint stability training, and perturbation training, balance training is used in this rehabilitation program for exercises pertaining to the first part of the definition (the ability to maintain a position). Accordingly, "balance training" exercises focus on awareness of posture and the position of the body in space with the aim of maintaining equilibrium without changing the base of support. Balance exercises include the double and single leg stance on even, flat surfaces, on a balance mat, on a wobble board, and on a trampoline. Three sensory systems contribute to the maintenance of balance: the visual, the vestibular, and the somatosensory (ie, proprioceptive) . Any of these systems may dominate, and all are context d e ~ e n d e n t . ~ ' Balance seems to affect both the injured leg and the uninjured leg in patients with ACL deficiency. Zatterstrom et a177found significant disturbance of balance in both the injured leg and the uninjured leg after ACL rupture compared with a reference group of normal subjects. Values of the uninjured leg were normalized after 3 months of training, but the injured leg still showed increased body sway. Normal balance parameters on the injured side were found at examination after 12 months and persisted up to 36 months posttraining.
Dynamic Joint Stability Training
'
..a,,-.;.,
:.
Balance Training
